linux-stable/drivers/interconnect/core.c

1149 lines
27 KiB
C
Raw Normal View History

// SPDX-License-Identifier: GPL-2.0
/*
* Interconnect framework core driver
*
* Copyright (c) 2017-2019, Linaro Ltd.
* Author: Georgi Djakov <georgi.djakov@linaro.org>
*/
#include <linux/debugfs.h>
#include <linux/device.h>
#include <linux/idr.h>
#include <linux/init.h>
#include <linux/interconnect.h>
#include <linux/interconnect-provider.h>
#include <linux/list.h>
#include <linux/mutex.h>
#include <linux/slab.h>
#include <linux/of.h>
#include <linux/overflow.h>
#include "internal.h"
#define CREATE_TRACE_POINTS
#include "trace.h"
static DEFINE_IDR(icc_idr);
static LIST_HEAD(icc_providers);
static int providers_count;
static bool synced_state;
static DEFINE_MUTEX(icc_lock);
interconnect: Fix locking for runpm vs reclaim For cases where icc_bw_set() can be called in callbaths that could deadlock against shrinker/reclaim, such as runpm resume, we need to decouple the icc locking. Introduce a new icc_bw_lock for cases where we need to serialize bw aggregation and update to decouple that from paths that require memory allocation such as node/link creation/ destruction. Fixes this lockdep splat: ====================================================== WARNING: possible circular locking dependency detected 6.2.0-rc8-debug+ #554 Not tainted ------------------------------------------------------ ring0/132 is trying to acquire lock: ffffff80871916d0 (&gmu->lock){+.+.}-{3:3}, at: a6xx_pm_resume+0xf0/0x234 but task is already holding lock: ffffffdb5aee57e8 (dma_fence_map){++++}-{0:0}, at: msm_job_run+0x68/0x150 which lock already depends on the new lock. the existing dependency chain (in reverse order) is: -> #4 (dma_fence_map){++++}-{0:0}: __dma_fence_might_wait+0x74/0xc0 dma_resv_lockdep+0x1f4/0x2f4 do_one_initcall+0x104/0x2bc kernel_init_freeable+0x344/0x34c kernel_init+0x30/0x134 ret_from_fork+0x10/0x20 -> #3 (mmu_notifier_invalidate_range_start){+.+.}-{0:0}: fs_reclaim_acquire+0x80/0xa8 slab_pre_alloc_hook.constprop.0+0x40/0x25c __kmem_cache_alloc_node+0x60/0x1cc __kmalloc+0xd8/0x100 topology_parse_cpu_capacity+0x8c/0x178 get_cpu_for_node+0x88/0xc4 parse_cluster+0x1b0/0x28c parse_cluster+0x8c/0x28c init_cpu_topology+0x168/0x188 smp_prepare_cpus+0x24/0xf8 kernel_init_freeable+0x18c/0x34c kernel_init+0x30/0x134 ret_from_fork+0x10/0x20 -> #2 (fs_reclaim){+.+.}-{0:0}: __fs_reclaim_acquire+0x3c/0x48 fs_reclaim_acquire+0x54/0xa8 slab_pre_alloc_hook.constprop.0+0x40/0x25c __kmem_cache_alloc_node+0x60/0x1cc __kmalloc+0xd8/0x100 kzalloc.constprop.0+0x14/0x20 icc_node_create_nolock+0x4c/0xc4 icc_node_create+0x38/0x58 qcom_icc_rpmh_probe+0x1b8/0x248 platform_probe+0x70/0xc4 really_probe+0x158/0x290 __driver_probe_device+0xc8/0xe0 driver_probe_device+0x44/0x100 __driver_attach+0xf8/0x108 bus_for_each_dev+0x78/0xc4 driver_attach+0x2c/0x38 bus_add_driver+0xd0/0x1d8 driver_register+0xbc/0xf8 __platform_driver_register+0x30/0x3c qnoc_driver_init+0x24/0x30 do_one_initcall+0x104/0x2bc kernel_init_freeable+0x344/0x34c kernel_init+0x30/0x134 ret_from_fork+0x10/0x20 -> #1 (icc_lock){+.+.}-{3:3}: __mutex_lock+0xcc/0x3c8 mutex_lock_nested+0x30/0x44 icc_set_bw+0x88/0x2b4 _set_opp_bw+0x8c/0xd8 _set_opp+0x19c/0x300 dev_pm_opp_set_opp+0x84/0x94 a6xx_gmu_resume+0x18c/0x804 a6xx_pm_resume+0xf8/0x234 adreno_runtime_resume+0x2c/0x38 pm_generic_runtime_resume+0x30/0x44 __rpm_callback+0x15c/0x174 rpm_callback+0x78/0x7c rpm_resume+0x318/0x524 __pm_runtime_resume+0x78/0xbc adreno_load_gpu+0xc4/0x17c msm_open+0x50/0x120 drm_file_alloc+0x17c/0x228 drm_open_helper+0x74/0x118 drm_open+0xa0/0x144 drm_stub_open+0xd4/0xe4 chrdev_open+0x1b8/0x1e4 do_dentry_open+0x2f8/0x38c vfs_open+0x34/0x40 path_openat+0x64c/0x7b4 do_filp_open+0x54/0xc4 do_sys_openat2+0x9c/0x100 do_sys_open+0x50/0x7c __arm64_sys_openat+0x28/0x34 invoke_syscall+0x8c/0x128 el0_svc_common.constprop.0+0xa0/0x11c do_el0_svc+0xac/0xbc el0_svc+0x48/0xa0 el0t_64_sync_handler+0xac/0x13c el0t_64_sync+0x190/0x194 -> #0 (&gmu->lock){+.+.}-{3:3}: __lock_acquire+0xe00/0x1060 lock_acquire+0x1e0/0x2f8 __mutex_lock+0xcc/0x3c8 mutex_lock_nested+0x30/0x44 a6xx_pm_resume+0xf0/0x234 adreno_runtime_resume+0x2c/0x38 pm_generic_runtime_resume+0x30/0x44 __rpm_callback+0x15c/0x174 rpm_callback+0x78/0x7c rpm_resume+0x318/0x524 __pm_runtime_resume+0x78/0xbc pm_runtime_get_sync.isra.0+0x14/0x20 msm_gpu_submit+0x58/0x178 msm_job_run+0x78/0x150 drm_sched_main+0x290/0x370 kthread+0xf0/0x100 ret_from_fork+0x10/0x20 other info that might help us debug this: Chain exists of: &gmu->lock --> mmu_notifier_invalidate_range_start --> dma_fence_map Possible unsafe locking scenario: CPU0 CPU1 ---- ---- lock(dma_fence_map); lock(mmu_notifier_invalidate_range_start); lock(dma_fence_map); lock(&gmu->lock); *** DEADLOCK *** 2 locks held by ring0/132: #0: ffffff8087191170 (&gpu->lock){+.+.}-{3:3}, at: msm_job_run+0x64/0x150 #1: ffffffdb5aee57e8 (dma_fence_map){++++}-{0:0}, at: msm_job_run+0x68/0x150 stack backtrace: CPU: 7 PID: 132 Comm: ring0 Not tainted 6.2.0-rc8-debug+ #554 Hardware name: Google Lazor (rev1 - 2) with LTE (DT) Call trace: dump_backtrace.part.0+0xb4/0xf8 show_stack+0x20/0x38 dump_stack_lvl+0x9c/0xd0 dump_stack+0x18/0x34 print_circular_bug+0x1b4/0x1f0 check_noncircular+0x78/0xac __lock_acquire+0xe00/0x1060 lock_acquire+0x1e0/0x2f8 __mutex_lock+0xcc/0x3c8 mutex_lock_nested+0x30/0x44 a6xx_pm_resume+0xf0/0x234 adreno_runtime_resume+0x2c/0x38 pm_generic_runtime_resume+0x30/0x44 __rpm_callback+0x15c/0x174 rpm_callback+0x78/0x7c rpm_resume+0x318/0x524 __pm_runtime_resume+0x78/0xbc pm_runtime_get_sync.isra.0+0x14/0x20 msm_gpu_submit+0x58/0x178 msm_job_run+0x78/0x150 drm_sched_main+0x290/0x370 kthread+0xf0/0x100 ret_from_fork+0x10/0x20 Signed-off-by: Rob Clark <robdclark@chromium.org> Link: https://lore.kernel.org/r/20230807171148.210181-7-robdclark@gmail.com Signed-off-by: Georgi Djakov <djakov@kernel.org>
2023-08-07 17:11:40 +00:00
static DEFINE_MUTEX(icc_bw_lock);
static struct dentry *icc_debugfs_dir;
static void icc_summary_show_one(struct seq_file *s, struct icc_node *n)
{
if (!n)
return;
seq_printf(s, "%-42s %12u %12u\n",
n->name, n->avg_bw, n->peak_bw);
}
static int icc_summary_show(struct seq_file *s, void *data)
{
struct icc_provider *provider;
seq_puts(s, " node tag avg peak\n");
seq_puts(s, "--------------------------------------------------------------------\n");
mutex_lock(&icc_lock);
list_for_each_entry(provider, &icc_providers, provider_list) {
struct icc_node *n;
list_for_each_entry(n, &provider->nodes, node_list) {
struct icc_req *r;
icc_summary_show_one(s, n);
hlist_for_each_entry(r, &n->req_list, req_node) {
u32 avg_bw = 0, peak_bw = 0;
if (!r->dev)
continue;
if (r->enabled) {
avg_bw = r->avg_bw;
peak_bw = r->peak_bw;
}
seq_printf(s, " %-27s %12u %12u %12u\n",
dev_name(r->dev), r->tag, avg_bw, peak_bw);
}
}
}
mutex_unlock(&icc_lock);
return 0;
}
DEFINE_SHOW_ATTRIBUTE(icc_summary);
static void icc_graph_show_link(struct seq_file *s, int level,
struct icc_node *n, struct icc_node *m)
{
seq_printf(s, "%s\"%d:%s\" -> \"%d:%s\"\n",
level == 2 ? "\t\t" : "\t",
n->id, n->name, m->id, m->name);
}
static void icc_graph_show_node(struct seq_file *s, struct icc_node *n)
{
seq_printf(s, "\t\t\"%d:%s\" [label=\"%d:%s",
n->id, n->name, n->id, n->name);
seq_printf(s, "\n\t\t\t|avg_bw=%ukBps", n->avg_bw);
seq_printf(s, "\n\t\t\t|peak_bw=%ukBps", n->peak_bw);
seq_puts(s, "\"]\n");
}
static int icc_graph_show(struct seq_file *s, void *data)
{
struct icc_provider *provider;
struct icc_node *n;
int cluster_index = 0;
int i;
seq_puts(s, "digraph {\n\trankdir = LR\n\tnode [shape = record]\n");
mutex_lock(&icc_lock);
/* draw providers as cluster subgraphs */
cluster_index = 0;
list_for_each_entry(provider, &icc_providers, provider_list) {
seq_printf(s, "\tsubgraph cluster_%d {\n", ++cluster_index);
if (provider->dev)
seq_printf(s, "\t\tlabel = \"%s\"\n",
dev_name(provider->dev));
/* draw nodes */
list_for_each_entry(n, &provider->nodes, node_list)
icc_graph_show_node(s, n);
/* draw internal links */
list_for_each_entry(n, &provider->nodes, node_list)
for (i = 0; i < n->num_links; ++i)
if (n->provider == n->links[i]->provider)
icc_graph_show_link(s, 2, n,
n->links[i]);
seq_puts(s, "\t}\n");
}
/* draw external links */
list_for_each_entry(provider, &icc_providers, provider_list)
list_for_each_entry(n, &provider->nodes, node_list)
for (i = 0; i < n->num_links; ++i)
if (n->provider != n->links[i]->provider)
icc_graph_show_link(s, 1, n,
n->links[i]);
mutex_unlock(&icc_lock);
seq_puts(s, "}");
return 0;
}
DEFINE_SHOW_ATTRIBUTE(icc_graph);
static struct icc_node *node_find(const int id)
{
return idr_find(&icc_idr, id);
}
static struct icc_node *node_find_by_name(const char *name)
{
struct icc_provider *provider;
struct icc_node *n;
list_for_each_entry(provider, &icc_providers, provider_list) {
list_for_each_entry(n, &provider->nodes, node_list) {
if (!strcmp(n->name, name))
return n;
}
}
return NULL;
}
static struct icc_path *path_init(struct device *dev, struct icc_node *dst,
ssize_t num_nodes)
{
struct icc_node *node = dst;
struct icc_path *path;
int i;
path = kzalloc(struct_size(path, reqs, num_nodes), GFP_KERNEL);
if (!path)
return ERR_PTR(-ENOMEM);
path->num_nodes = num_nodes;
interconnect: Don't access req_list while it's being manipulated The icc_lock mutex was split into separate icc_lock and icc_bw_lock mutexes in [1] to avoid lockdep splats. However, this didn't adequately protect access to icc_node::req_list. The icc_set_bw() function will eventually iterate over req_list while only holding icc_bw_lock, but req_list can be modified while only holding icc_lock. This causes races between icc_set_bw(), of_icc_get(), and icc_put(). Example A: CPU0 CPU1 ---- ---- icc_set_bw(path_a) mutex_lock(&icc_bw_lock); icc_put(path_b) mutex_lock(&icc_lock); aggregate_requests() hlist_for_each_entry(r, ... hlist_del(... <r = invalid pointer> Example B: CPU0 CPU1 ---- ---- icc_set_bw(path_a) mutex_lock(&icc_bw_lock); path_b = of_icc_get() of_icc_get_by_index() mutex_lock(&icc_lock); path_find() path_init() aggregate_requests() hlist_for_each_entry(r, ... hlist_add_head(... <r = invalid pointer> Fix this by ensuring icc_bw_lock is always held before manipulating icc_node::req_list. The additional places icc_bw_lock is held don't perform any memory allocations, so we should still be safe from the original lockdep splats that motivated the separate locks. [1] commit af42269c3523 ("interconnect: Fix locking for runpm vs reclaim") Signed-off-by: Mike Tipton <quic_mdtipton@quicinc.com> Fixes: af42269c3523 ("interconnect: Fix locking for runpm vs reclaim") Reviewed-by: Rob Clark <robdclark@chromium.org> Link: https://lore.kernel.org/r/20240305225652.22872-1-quic_mdtipton@quicinc.com Signed-off-by: Georgi Djakov <djakov@kernel.org>
2024-03-05 22:56:52 +00:00
mutex_lock(&icc_bw_lock);
for (i = num_nodes - 1; i >= 0; i--) {
node->provider->users++;
hlist_add_head(&path->reqs[i].req_node, &node->req_list);
path->reqs[i].node = node;
path->reqs[i].dev = dev;
path->reqs[i].enabled = true;
/* reference to previous node was saved during path traversal */
node = node->reverse;
}
interconnect: Don't access req_list while it's being manipulated The icc_lock mutex was split into separate icc_lock and icc_bw_lock mutexes in [1] to avoid lockdep splats. However, this didn't adequately protect access to icc_node::req_list. The icc_set_bw() function will eventually iterate over req_list while only holding icc_bw_lock, but req_list can be modified while only holding icc_lock. This causes races between icc_set_bw(), of_icc_get(), and icc_put(). Example A: CPU0 CPU1 ---- ---- icc_set_bw(path_a) mutex_lock(&icc_bw_lock); icc_put(path_b) mutex_lock(&icc_lock); aggregate_requests() hlist_for_each_entry(r, ... hlist_del(... <r = invalid pointer> Example B: CPU0 CPU1 ---- ---- icc_set_bw(path_a) mutex_lock(&icc_bw_lock); path_b = of_icc_get() of_icc_get_by_index() mutex_lock(&icc_lock); path_find() path_init() aggregate_requests() hlist_for_each_entry(r, ... hlist_add_head(... <r = invalid pointer> Fix this by ensuring icc_bw_lock is always held before manipulating icc_node::req_list. The additional places icc_bw_lock is held don't perform any memory allocations, so we should still be safe from the original lockdep splats that motivated the separate locks. [1] commit af42269c3523 ("interconnect: Fix locking for runpm vs reclaim") Signed-off-by: Mike Tipton <quic_mdtipton@quicinc.com> Fixes: af42269c3523 ("interconnect: Fix locking for runpm vs reclaim") Reviewed-by: Rob Clark <robdclark@chromium.org> Link: https://lore.kernel.org/r/20240305225652.22872-1-quic_mdtipton@quicinc.com Signed-off-by: Georgi Djakov <djakov@kernel.org>
2024-03-05 22:56:52 +00:00
mutex_unlock(&icc_bw_lock);
return path;
}
static struct icc_path *path_find(struct device *dev, struct icc_node *src,
struct icc_node *dst)
{
struct icc_path *path = ERR_PTR(-EPROBE_DEFER);
struct icc_node *n, *node = NULL;
struct list_head traverse_list;
struct list_head edge_list;
struct list_head visited_list;
size_t i, depth = 1;
bool found = false;
INIT_LIST_HEAD(&traverse_list);
INIT_LIST_HEAD(&edge_list);
INIT_LIST_HEAD(&visited_list);
list_add(&src->search_list, &traverse_list);
src->reverse = NULL;
do {
list_for_each_entry_safe(node, n, &traverse_list, search_list) {
if (node == dst) {
found = true;
list_splice_init(&edge_list, &visited_list);
list_splice_init(&traverse_list, &visited_list);
break;
}
for (i = 0; i < node->num_links; i++) {
struct icc_node *tmp = node->links[i];
if (!tmp) {
path = ERR_PTR(-ENOENT);
goto out;
}
if (tmp->is_traversed)
continue;
tmp->is_traversed = true;
tmp->reverse = node;
list_add_tail(&tmp->search_list, &edge_list);
}
}
if (found)
break;
list_splice_init(&traverse_list, &visited_list);
list_splice_init(&edge_list, &traverse_list);
/* count the hops including the source */
depth++;
} while (!list_empty(&traverse_list));
out:
/* reset the traversed state */
list_for_each_entry_reverse(n, &visited_list, search_list)
n->is_traversed = false;
if (found)
path = path_init(dev, dst, depth);
return path;
}
/*
* We want the path to honor all bandwidth requests, so the average and peak
* bandwidth requirements from each consumer are aggregated at each node.
* The aggregation is platform specific, so each platform can customize it by
* implementing its own aggregate() function.
*/
static int aggregate_requests(struct icc_node *node)
{
struct icc_provider *p = node->provider;
struct icc_req *r;
u32 avg_bw, peak_bw;
node->avg_bw = 0;
node->peak_bw = 0;
if (p->pre_aggregate)
p->pre_aggregate(node);
hlist_for_each_entry(r, &node->req_list, req_node) {
if (r->enabled) {
avg_bw = r->avg_bw;
peak_bw = r->peak_bw;
} else {
avg_bw = 0;
peak_bw = 0;
}
p->aggregate(node, r->tag, avg_bw, peak_bw,
&node->avg_bw, &node->peak_bw);
/* during boot use the initial bandwidth as a floor value */
if (!synced_state) {
node->avg_bw = max(node->avg_bw, node->init_avg);
node->peak_bw = max(node->peak_bw, node->init_peak);
}
}
return 0;
}
static int apply_constraints(struct icc_path *path)
{
struct icc_node *next, *prev = NULL;
struct icc_provider *p;
int ret = -EINVAL;
int i;
for (i = 0; i < path->num_nodes; i++) {
next = path->reqs[i].node;
p = next->provider;
/* both endpoints should be valid master-slave pairs */
if (!prev || (p != prev->provider && !p->inter_set)) {
prev = next;
continue;
}
/* set the constraints */
ret = p->set(prev, next);
if (ret)
goto out;
prev = next;
}
out:
return ret;
}
int icc_std_aggregate(struct icc_node *node, u32 tag, u32 avg_bw,
u32 peak_bw, u32 *agg_avg, u32 *agg_peak)
{
*agg_avg += avg_bw;
*agg_peak = max(*agg_peak, peak_bw);
return 0;
}
EXPORT_SYMBOL_GPL(icc_std_aggregate);
/* of_icc_xlate_onecell() - Translate function using a single index.
* @spec: OF phandle args to map into an interconnect node.
* @data: private data (pointer to struct icc_onecell_data)
*
* This is a generic translate function that can be used to model simple
* interconnect providers that have one device tree node and provide
* multiple interconnect nodes. A single cell is used as an index into
* an array of icc nodes specified in the icc_onecell_data struct when
* registering the provider.
*/
struct icc_node *of_icc_xlate_onecell(const struct of_phandle_args *spec,
void *data)
{
struct icc_onecell_data *icc_data = data;
unsigned int idx = spec->args[0];
if (idx >= icc_data->num_nodes) {
pr_err("%s: invalid index %u\n", __func__, idx);
return ERR_PTR(-EINVAL);
}
return icc_data->nodes[idx];
}
EXPORT_SYMBOL_GPL(of_icc_xlate_onecell);
/**
* of_icc_get_from_provider() - Look-up interconnect node
* @spec: OF phandle args to use for look-up
*
* Looks for interconnect provider under the node specified by @spec and if
* found, uses xlate function of the provider to map phandle args to node.
*
* Returns a valid pointer to struct icc_node_data on success or ERR_PTR()
* on failure.
*/
struct icc_node_data *of_icc_get_from_provider(const struct of_phandle_args *spec)
{
struct icc_node *node = ERR_PTR(-EPROBE_DEFER);
struct icc_node_data *data = NULL;
struct icc_provider *provider;
if (!spec)
return ERR_PTR(-EINVAL);
mutex_lock(&icc_lock);
list_for_each_entry(provider, &icc_providers, provider_list) {
if (provider->dev->of_node == spec->np) {
if (provider->xlate_extended) {
data = provider->xlate_extended(spec, provider->data);
if (!IS_ERR(data)) {
node = data->node;
break;
}
} else {
node = provider->xlate(spec, provider->data);
if (!IS_ERR(node))
break;
}
}
}
mutex_unlock(&icc_lock);
if (!node)
return ERR_PTR(-EINVAL);
if (IS_ERR(node))
return ERR_CAST(node);
if (!data) {
data = kzalloc(sizeof(*data), GFP_KERNEL);
if (!data)
return ERR_PTR(-ENOMEM);
data->node = node;
}
return data;
}
EXPORT_SYMBOL_GPL(of_icc_get_from_provider);
static void devm_icc_release(struct device *dev, void *res)
{
icc_put(*(struct icc_path **)res);
}
struct icc_path *devm_of_icc_get(struct device *dev, const char *name)
{
struct icc_path **ptr, *path;
ptr = devres_alloc(devm_icc_release, sizeof(*ptr), GFP_KERNEL);
if (!ptr)
return ERR_PTR(-ENOMEM);
path = of_icc_get(dev, name);
if (!IS_ERR(path)) {
*ptr = path;
devres_add(dev, ptr);
} else {
devres_free(ptr);
}
return path;
}
EXPORT_SYMBOL_GPL(devm_of_icc_get);
/**
* of_icc_get_by_index() - get a path handle from a DT node based on index
* @dev: device pointer for the consumer device
* @idx: interconnect path index
*
* This function will search for a path between two endpoints and return an
* icc_path handle on success. Use icc_put() to release constraints when they
* are not needed anymore.
* If the interconnect API is disabled, NULL is returned and the consumer
* drivers will still build. Drivers are free to handle this specifically,
* but they don't have to.
*
* Return: icc_path pointer on success or ERR_PTR() on error. NULL is returned
* when the API is disabled or the "interconnects" DT property is missing.
*/
struct icc_path *of_icc_get_by_index(struct device *dev, int idx)
{
struct icc_path *path;
struct icc_node_data *src_data, *dst_data;
struct device_node *np;
struct of_phandle_args src_args, dst_args;
int ret;
if (!dev || !dev->of_node)
return ERR_PTR(-ENODEV);
np = dev->of_node;
/*
* When the consumer DT node do not have "interconnects" property
* return a NULL path to skip setting constraints.
*/
if (!of_property_present(np, "interconnects"))
return NULL;
/*
* We use a combination of phandle and specifier for endpoint. For now
* lets support only global ids and extend this in the future if needed
* without breaking DT compatibility.
*/
ret = of_parse_phandle_with_args(np, "interconnects",
"#interconnect-cells", idx * 2,
&src_args);
if (ret)
return ERR_PTR(ret);
of_node_put(src_args.np);
ret = of_parse_phandle_with_args(np, "interconnects",
"#interconnect-cells", idx * 2 + 1,
&dst_args);
if (ret)
return ERR_PTR(ret);
of_node_put(dst_args.np);
src_data = of_icc_get_from_provider(&src_args);
if (IS_ERR(src_data)) {
dev_err_probe(dev, PTR_ERR(src_data), "error finding src node\n");
return ERR_CAST(src_data);
}
dst_data = of_icc_get_from_provider(&dst_args);
if (IS_ERR(dst_data)) {
dev_err_probe(dev, PTR_ERR(dst_data), "error finding dst node\n");
kfree(src_data);
return ERR_CAST(dst_data);
}
mutex_lock(&icc_lock);
path = path_find(dev, src_data->node, dst_data->node);
mutex_unlock(&icc_lock);
if (IS_ERR(path)) {
dev_err(dev, "%s: invalid path=%ld\n", __func__, PTR_ERR(path));
goto free_icc_data;
}
if (src_data->tag && src_data->tag == dst_data->tag)
icc_set_tag(path, src_data->tag);
path->name = kasprintf(GFP_KERNEL, "%s-%s",
src_data->node->name, dst_data->node->name);
if (!path->name) {
kfree(path);
path = ERR_PTR(-ENOMEM);
}
free_icc_data:
kfree(src_data);
kfree(dst_data);
return path;
}
EXPORT_SYMBOL_GPL(of_icc_get_by_index);
/**
* of_icc_get() - get a path handle from a DT node based on name
* @dev: device pointer for the consumer device
* @name: interconnect path name
*
* This function will search for a path between two endpoints and return an
* icc_path handle on success. Use icc_put() to release constraints when they
* are not needed anymore.
* If the interconnect API is disabled, NULL is returned and the consumer
* drivers will still build. Drivers are free to handle this specifically,
* but they don't have to.
*
* Return: icc_path pointer on success or ERR_PTR() on error. NULL is returned
* when the API is disabled or the "interconnects" DT property is missing.
*/
struct icc_path *of_icc_get(struct device *dev, const char *name)
{
struct device_node *np;
int idx = 0;
if (!dev || !dev->of_node)
return ERR_PTR(-ENODEV);
np = dev->of_node;
/*
* When the consumer DT node do not have "interconnects" property
* return a NULL path to skip setting constraints.
*/
if (!of_property_present(np, "interconnects"))
return NULL;
/*
* We use a combination of phandle and specifier for endpoint. For now
* lets support only global ids and extend this in the future if needed
* without breaking DT compatibility.
*/
if (name) {
idx = of_property_match_string(np, "interconnect-names", name);
if (idx < 0)
return ERR_PTR(idx);
}
return of_icc_get_by_index(dev, idx);
}
EXPORT_SYMBOL_GPL(of_icc_get);
/**
* icc_get() - get a path handle between two endpoints
* @dev: device pointer for the consumer device
* @src: source node name
* @dst: destination node name
*
* This function will search for a path between two endpoints and return an
* icc_path handle on success. Use icc_put() to release constraints when they
* are not needed anymore.
*
* Return: icc_path pointer on success or ERR_PTR() on error. NULL is returned
* when the API is disabled.
*/
struct icc_path *icc_get(struct device *dev, const char *src, const char *dst)
{
struct icc_node *src_node, *dst_node;
struct icc_path *path = ERR_PTR(-EPROBE_DEFER);
mutex_lock(&icc_lock);
src_node = node_find_by_name(src);
if (!src_node) {
dev_err(dev, "%s: invalid src=%s\n", __func__, src);
goto out;
}
dst_node = node_find_by_name(dst);
if (!dst_node) {
dev_err(dev, "%s: invalid dst=%s\n", __func__, dst);
goto out;
}
path = path_find(dev, src_node, dst_node);
if (IS_ERR(path)) {
dev_err(dev, "%s: invalid path=%ld\n", __func__, PTR_ERR(path));
goto out;
}
path->name = kasprintf(GFP_KERNEL, "%s-%s", src_node->name, dst_node->name);
if (!path->name) {
kfree(path);
path = ERR_PTR(-ENOMEM);
}
out:
mutex_unlock(&icc_lock);
return path;
}
/**
* icc_set_tag() - set an optional tag on a path
* @path: the path we want to tag
* @tag: the tag value
*
* This function allows consumers to append a tag to the requests associated
* with a path, so that a different aggregation could be done based on this tag.
*/
void icc_set_tag(struct icc_path *path, u32 tag)
{
int i;
if (!path)
return;
mutex_lock(&icc_lock);
for (i = 0; i < path->num_nodes; i++)
path->reqs[i].tag = tag;
mutex_unlock(&icc_lock);
}
EXPORT_SYMBOL_GPL(icc_set_tag);
/**
* icc_get_name() - Get name of the icc path
* @path: interconnect path
*
* This function is used by an interconnect consumer to get the name of the icc
* path.
*
* Returns a valid pointer on success, or NULL otherwise.
*/
const char *icc_get_name(struct icc_path *path)
{
if (!path)
return NULL;
return path->name;
}
EXPORT_SYMBOL_GPL(icc_get_name);
/**
* icc_set_bw() - set bandwidth constraints on an interconnect path
* @path: interconnect path
* @avg_bw: average bandwidth in kilobytes per second
* @peak_bw: peak bandwidth in kilobytes per second
*
* This function is used by an interconnect consumer to express its own needs
* in terms of bandwidth for a previously requested path between two endpoints.
* The requests are aggregated and each node is updated accordingly. The entire
* path is locked by a mutex to ensure that the set() is completed.
* The @path can be NULL when the "interconnects" DT properties is missing,
* which will mean that no constraints will be set.
*
* Returns 0 on success, or an appropriate error code otherwise.
*/
int icc_set_bw(struct icc_path *path, u32 avg_bw, u32 peak_bw)
{
struct icc_node *node;
u32 old_avg, old_peak;
size_t i;
int ret;
if (!path)
return 0;
if (WARN_ON(IS_ERR(path) || !path->num_nodes))
return -EINVAL;
interconnect: Fix locking for runpm vs reclaim For cases where icc_bw_set() can be called in callbaths that could deadlock against shrinker/reclaim, such as runpm resume, we need to decouple the icc locking. Introduce a new icc_bw_lock for cases where we need to serialize bw aggregation and update to decouple that from paths that require memory allocation such as node/link creation/ destruction. Fixes this lockdep splat: ====================================================== WARNING: possible circular locking dependency detected 6.2.0-rc8-debug+ #554 Not tainted ------------------------------------------------------ ring0/132 is trying to acquire lock: ffffff80871916d0 (&gmu->lock){+.+.}-{3:3}, at: a6xx_pm_resume+0xf0/0x234 but task is already holding lock: ffffffdb5aee57e8 (dma_fence_map){++++}-{0:0}, at: msm_job_run+0x68/0x150 which lock already depends on the new lock. the existing dependency chain (in reverse order) is: -> #4 (dma_fence_map){++++}-{0:0}: __dma_fence_might_wait+0x74/0xc0 dma_resv_lockdep+0x1f4/0x2f4 do_one_initcall+0x104/0x2bc kernel_init_freeable+0x344/0x34c kernel_init+0x30/0x134 ret_from_fork+0x10/0x20 -> #3 (mmu_notifier_invalidate_range_start){+.+.}-{0:0}: fs_reclaim_acquire+0x80/0xa8 slab_pre_alloc_hook.constprop.0+0x40/0x25c __kmem_cache_alloc_node+0x60/0x1cc __kmalloc+0xd8/0x100 topology_parse_cpu_capacity+0x8c/0x178 get_cpu_for_node+0x88/0xc4 parse_cluster+0x1b0/0x28c parse_cluster+0x8c/0x28c init_cpu_topology+0x168/0x188 smp_prepare_cpus+0x24/0xf8 kernel_init_freeable+0x18c/0x34c kernel_init+0x30/0x134 ret_from_fork+0x10/0x20 -> #2 (fs_reclaim){+.+.}-{0:0}: __fs_reclaim_acquire+0x3c/0x48 fs_reclaim_acquire+0x54/0xa8 slab_pre_alloc_hook.constprop.0+0x40/0x25c __kmem_cache_alloc_node+0x60/0x1cc __kmalloc+0xd8/0x100 kzalloc.constprop.0+0x14/0x20 icc_node_create_nolock+0x4c/0xc4 icc_node_create+0x38/0x58 qcom_icc_rpmh_probe+0x1b8/0x248 platform_probe+0x70/0xc4 really_probe+0x158/0x290 __driver_probe_device+0xc8/0xe0 driver_probe_device+0x44/0x100 __driver_attach+0xf8/0x108 bus_for_each_dev+0x78/0xc4 driver_attach+0x2c/0x38 bus_add_driver+0xd0/0x1d8 driver_register+0xbc/0xf8 __platform_driver_register+0x30/0x3c qnoc_driver_init+0x24/0x30 do_one_initcall+0x104/0x2bc kernel_init_freeable+0x344/0x34c kernel_init+0x30/0x134 ret_from_fork+0x10/0x20 -> #1 (icc_lock){+.+.}-{3:3}: __mutex_lock+0xcc/0x3c8 mutex_lock_nested+0x30/0x44 icc_set_bw+0x88/0x2b4 _set_opp_bw+0x8c/0xd8 _set_opp+0x19c/0x300 dev_pm_opp_set_opp+0x84/0x94 a6xx_gmu_resume+0x18c/0x804 a6xx_pm_resume+0xf8/0x234 adreno_runtime_resume+0x2c/0x38 pm_generic_runtime_resume+0x30/0x44 __rpm_callback+0x15c/0x174 rpm_callback+0x78/0x7c rpm_resume+0x318/0x524 __pm_runtime_resume+0x78/0xbc adreno_load_gpu+0xc4/0x17c msm_open+0x50/0x120 drm_file_alloc+0x17c/0x228 drm_open_helper+0x74/0x118 drm_open+0xa0/0x144 drm_stub_open+0xd4/0xe4 chrdev_open+0x1b8/0x1e4 do_dentry_open+0x2f8/0x38c vfs_open+0x34/0x40 path_openat+0x64c/0x7b4 do_filp_open+0x54/0xc4 do_sys_openat2+0x9c/0x100 do_sys_open+0x50/0x7c __arm64_sys_openat+0x28/0x34 invoke_syscall+0x8c/0x128 el0_svc_common.constprop.0+0xa0/0x11c do_el0_svc+0xac/0xbc el0_svc+0x48/0xa0 el0t_64_sync_handler+0xac/0x13c el0t_64_sync+0x190/0x194 -> #0 (&gmu->lock){+.+.}-{3:3}: __lock_acquire+0xe00/0x1060 lock_acquire+0x1e0/0x2f8 __mutex_lock+0xcc/0x3c8 mutex_lock_nested+0x30/0x44 a6xx_pm_resume+0xf0/0x234 adreno_runtime_resume+0x2c/0x38 pm_generic_runtime_resume+0x30/0x44 __rpm_callback+0x15c/0x174 rpm_callback+0x78/0x7c rpm_resume+0x318/0x524 __pm_runtime_resume+0x78/0xbc pm_runtime_get_sync.isra.0+0x14/0x20 msm_gpu_submit+0x58/0x178 msm_job_run+0x78/0x150 drm_sched_main+0x290/0x370 kthread+0xf0/0x100 ret_from_fork+0x10/0x20 other info that might help us debug this: Chain exists of: &gmu->lock --> mmu_notifier_invalidate_range_start --> dma_fence_map Possible unsafe locking scenario: CPU0 CPU1 ---- ---- lock(dma_fence_map); lock(mmu_notifier_invalidate_range_start); lock(dma_fence_map); lock(&gmu->lock); *** DEADLOCK *** 2 locks held by ring0/132: #0: ffffff8087191170 (&gpu->lock){+.+.}-{3:3}, at: msm_job_run+0x64/0x150 #1: ffffffdb5aee57e8 (dma_fence_map){++++}-{0:0}, at: msm_job_run+0x68/0x150 stack backtrace: CPU: 7 PID: 132 Comm: ring0 Not tainted 6.2.0-rc8-debug+ #554 Hardware name: Google Lazor (rev1 - 2) with LTE (DT) Call trace: dump_backtrace.part.0+0xb4/0xf8 show_stack+0x20/0x38 dump_stack_lvl+0x9c/0xd0 dump_stack+0x18/0x34 print_circular_bug+0x1b4/0x1f0 check_noncircular+0x78/0xac __lock_acquire+0xe00/0x1060 lock_acquire+0x1e0/0x2f8 __mutex_lock+0xcc/0x3c8 mutex_lock_nested+0x30/0x44 a6xx_pm_resume+0xf0/0x234 adreno_runtime_resume+0x2c/0x38 pm_generic_runtime_resume+0x30/0x44 __rpm_callback+0x15c/0x174 rpm_callback+0x78/0x7c rpm_resume+0x318/0x524 __pm_runtime_resume+0x78/0xbc pm_runtime_get_sync.isra.0+0x14/0x20 msm_gpu_submit+0x58/0x178 msm_job_run+0x78/0x150 drm_sched_main+0x290/0x370 kthread+0xf0/0x100 ret_from_fork+0x10/0x20 Signed-off-by: Rob Clark <robdclark@chromium.org> Link: https://lore.kernel.org/r/20230807171148.210181-7-robdclark@gmail.com Signed-off-by: Georgi Djakov <djakov@kernel.org>
2023-08-07 17:11:40 +00:00
mutex_lock(&icc_bw_lock);
old_avg = path->reqs[0].avg_bw;
old_peak = path->reqs[0].peak_bw;
for (i = 0; i < path->num_nodes; i++) {
node = path->reqs[i].node;
/* update the consumer request for this path */
path->reqs[i].avg_bw = avg_bw;
path->reqs[i].peak_bw = peak_bw;
/* aggregate requests for this node */
aggregate_requests(node);
trace_icc_set_bw(path, node, i, avg_bw, peak_bw);
}
ret = apply_constraints(path);
if (ret) {
pr_debug("interconnect: error applying constraints (%d)\n",
ret);
for (i = 0; i < path->num_nodes; i++) {
node = path->reqs[i].node;
path->reqs[i].avg_bw = old_avg;
path->reqs[i].peak_bw = old_peak;
aggregate_requests(node);
}
apply_constraints(path);
}
interconnect: Fix locking for runpm vs reclaim For cases where icc_bw_set() can be called in callbaths that could deadlock against shrinker/reclaim, such as runpm resume, we need to decouple the icc locking. Introduce a new icc_bw_lock for cases where we need to serialize bw aggregation and update to decouple that from paths that require memory allocation such as node/link creation/ destruction. Fixes this lockdep splat: ====================================================== WARNING: possible circular locking dependency detected 6.2.0-rc8-debug+ #554 Not tainted ------------------------------------------------------ ring0/132 is trying to acquire lock: ffffff80871916d0 (&gmu->lock){+.+.}-{3:3}, at: a6xx_pm_resume+0xf0/0x234 but task is already holding lock: ffffffdb5aee57e8 (dma_fence_map){++++}-{0:0}, at: msm_job_run+0x68/0x150 which lock already depends on the new lock. the existing dependency chain (in reverse order) is: -> #4 (dma_fence_map){++++}-{0:0}: __dma_fence_might_wait+0x74/0xc0 dma_resv_lockdep+0x1f4/0x2f4 do_one_initcall+0x104/0x2bc kernel_init_freeable+0x344/0x34c kernel_init+0x30/0x134 ret_from_fork+0x10/0x20 -> #3 (mmu_notifier_invalidate_range_start){+.+.}-{0:0}: fs_reclaim_acquire+0x80/0xa8 slab_pre_alloc_hook.constprop.0+0x40/0x25c __kmem_cache_alloc_node+0x60/0x1cc __kmalloc+0xd8/0x100 topology_parse_cpu_capacity+0x8c/0x178 get_cpu_for_node+0x88/0xc4 parse_cluster+0x1b0/0x28c parse_cluster+0x8c/0x28c init_cpu_topology+0x168/0x188 smp_prepare_cpus+0x24/0xf8 kernel_init_freeable+0x18c/0x34c kernel_init+0x30/0x134 ret_from_fork+0x10/0x20 -> #2 (fs_reclaim){+.+.}-{0:0}: __fs_reclaim_acquire+0x3c/0x48 fs_reclaim_acquire+0x54/0xa8 slab_pre_alloc_hook.constprop.0+0x40/0x25c __kmem_cache_alloc_node+0x60/0x1cc __kmalloc+0xd8/0x100 kzalloc.constprop.0+0x14/0x20 icc_node_create_nolock+0x4c/0xc4 icc_node_create+0x38/0x58 qcom_icc_rpmh_probe+0x1b8/0x248 platform_probe+0x70/0xc4 really_probe+0x158/0x290 __driver_probe_device+0xc8/0xe0 driver_probe_device+0x44/0x100 __driver_attach+0xf8/0x108 bus_for_each_dev+0x78/0xc4 driver_attach+0x2c/0x38 bus_add_driver+0xd0/0x1d8 driver_register+0xbc/0xf8 __platform_driver_register+0x30/0x3c qnoc_driver_init+0x24/0x30 do_one_initcall+0x104/0x2bc kernel_init_freeable+0x344/0x34c kernel_init+0x30/0x134 ret_from_fork+0x10/0x20 -> #1 (icc_lock){+.+.}-{3:3}: __mutex_lock+0xcc/0x3c8 mutex_lock_nested+0x30/0x44 icc_set_bw+0x88/0x2b4 _set_opp_bw+0x8c/0xd8 _set_opp+0x19c/0x300 dev_pm_opp_set_opp+0x84/0x94 a6xx_gmu_resume+0x18c/0x804 a6xx_pm_resume+0xf8/0x234 adreno_runtime_resume+0x2c/0x38 pm_generic_runtime_resume+0x30/0x44 __rpm_callback+0x15c/0x174 rpm_callback+0x78/0x7c rpm_resume+0x318/0x524 __pm_runtime_resume+0x78/0xbc adreno_load_gpu+0xc4/0x17c msm_open+0x50/0x120 drm_file_alloc+0x17c/0x228 drm_open_helper+0x74/0x118 drm_open+0xa0/0x144 drm_stub_open+0xd4/0xe4 chrdev_open+0x1b8/0x1e4 do_dentry_open+0x2f8/0x38c vfs_open+0x34/0x40 path_openat+0x64c/0x7b4 do_filp_open+0x54/0xc4 do_sys_openat2+0x9c/0x100 do_sys_open+0x50/0x7c __arm64_sys_openat+0x28/0x34 invoke_syscall+0x8c/0x128 el0_svc_common.constprop.0+0xa0/0x11c do_el0_svc+0xac/0xbc el0_svc+0x48/0xa0 el0t_64_sync_handler+0xac/0x13c el0t_64_sync+0x190/0x194 -> #0 (&gmu->lock){+.+.}-{3:3}: __lock_acquire+0xe00/0x1060 lock_acquire+0x1e0/0x2f8 __mutex_lock+0xcc/0x3c8 mutex_lock_nested+0x30/0x44 a6xx_pm_resume+0xf0/0x234 adreno_runtime_resume+0x2c/0x38 pm_generic_runtime_resume+0x30/0x44 __rpm_callback+0x15c/0x174 rpm_callback+0x78/0x7c rpm_resume+0x318/0x524 __pm_runtime_resume+0x78/0xbc pm_runtime_get_sync.isra.0+0x14/0x20 msm_gpu_submit+0x58/0x178 msm_job_run+0x78/0x150 drm_sched_main+0x290/0x370 kthread+0xf0/0x100 ret_from_fork+0x10/0x20 other info that might help us debug this: Chain exists of: &gmu->lock --> mmu_notifier_invalidate_range_start --> dma_fence_map Possible unsafe locking scenario: CPU0 CPU1 ---- ---- lock(dma_fence_map); lock(mmu_notifier_invalidate_range_start); lock(dma_fence_map); lock(&gmu->lock); *** DEADLOCK *** 2 locks held by ring0/132: #0: ffffff8087191170 (&gpu->lock){+.+.}-{3:3}, at: msm_job_run+0x64/0x150 #1: ffffffdb5aee57e8 (dma_fence_map){++++}-{0:0}, at: msm_job_run+0x68/0x150 stack backtrace: CPU: 7 PID: 132 Comm: ring0 Not tainted 6.2.0-rc8-debug+ #554 Hardware name: Google Lazor (rev1 - 2) with LTE (DT) Call trace: dump_backtrace.part.0+0xb4/0xf8 show_stack+0x20/0x38 dump_stack_lvl+0x9c/0xd0 dump_stack+0x18/0x34 print_circular_bug+0x1b4/0x1f0 check_noncircular+0x78/0xac __lock_acquire+0xe00/0x1060 lock_acquire+0x1e0/0x2f8 __mutex_lock+0xcc/0x3c8 mutex_lock_nested+0x30/0x44 a6xx_pm_resume+0xf0/0x234 adreno_runtime_resume+0x2c/0x38 pm_generic_runtime_resume+0x30/0x44 __rpm_callback+0x15c/0x174 rpm_callback+0x78/0x7c rpm_resume+0x318/0x524 __pm_runtime_resume+0x78/0xbc pm_runtime_get_sync.isra.0+0x14/0x20 msm_gpu_submit+0x58/0x178 msm_job_run+0x78/0x150 drm_sched_main+0x290/0x370 kthread+0xf0/0x100 ret_from_fork+0x10/0x20 Signed-off-by: Rob Clark <robdclark@chromium.org> Link: https://lore.kernel.org/r/20230807171148.210181-7-robdclark@gmail.com Signed-off-by: Georgi Djakov <djakov@kernel.org>
2023-08-07 17:11:40 +00:00
mutex_unlock(&icc_bw_lock);
trace_icc_set_bw_end(path, ret);
return ret;
}
EXPORT_SYMBOL_GPL(icc_set_bw);
static int __icc_enable(struct icc_path *path, bool enable)
{
int i;
if (!path)
return 0;
if (WARN_ON(IS_ERR(path) || !path->num_nodes))
return -EINVAL;
mutex_lock(&icc_lock);
for (i = 0; i < path->num_nodes; i++)
path->reqs[i].enabled = enable;
mutex_unlock(&icc_lock);
return icc_set_bw(path, path->reqs[0].avg_bw,
path->reqs[0].peak_bw);
}
int icc_enable(struct icc_path *path)
{
return __icc_enable(path, true);
}
EXPORT_SYMBOL_GPL(icc_enable);
int icc_disable(struct icc_path *path)
{
return __icc_enable(path, false);
}
EXPORT_SYMBOL_GPL(icc_disable);
/**
* icc_put() - release the reference to the icc_path
* @path: interconnect path
*
* Use this function to release the constraints on a path when the path is
* no longer needed. The constraints will be re-aggregated.
*/
void icc_put(struct icc_path *path)
{
struct icc_node *node;
size_t i;
int ret;
if (!path || WARN_ON(IS_ERR(path)))
return;
ret = icc_set_bw(path, 0, 0);
if (ret)
pr_err("%s: error (%d)\n", __func__, ret);
mutex_lock(&icc_lock);
interconnect: Don't access req_list while it's being manipulated The icc_lock mutex was split into separate icc_lock and icc_bw_lock mutexes in [1] to avoid lockdep splats. However, this didn't adequately protect access to icc_node::req_list. The icc_set_bw() function will eventually iterate over req_list while only holding icc_bw_lock, but req_list can be modified while only holding icc_lock. This causes races between icc_set_bw(), of_icc_get(), and icc_put(). Example A: CPU0 CPU1 ---- ---- icc_set_bw(path_a) mutex_lock(&icc_bw_lock); icc_put(path_b) mutex_lock(&icc_lock); aggregate_requests() hlist_for_each_entry(r, ... hlist_del(... <r = invalid pointer> Example B: CPU0 CPU1 ---- ---- icc_set_bw(path_a) mutex_lock(&icc_bw_lock); path_b = of_icc_get() of_icc_get_by_index() mutex_lock(&icc_lock); path_find() path_init() aggregate_requests() hlist_for_each_entry(r, ... hlist_add_head(... <r = invalid pointer> Fix this by ensuring icc_bw_lock is always held before manipulating icc_node::req_list. The additional places icc_bw_lock is held don't perform any memory allocations, so we should still be safe from the original lockdep splats that motivated the separate locks. [1] commit af42269c3523 ("interconnect: Fix locking for runpm vs reclaim") Signed-off-by: Mike Tipton <quic_mdtipton@quicinc.com> Fixes: af42269c3523 ("interconnect: Fix locking for runpm vs reclaim") Reviewed-by: Rob Clark <robdclark@chromium.org> Link: https://lore.kernel.org/r/20240305225652.22872-1-quic_mdtipton@quicinc.com Signed-off-by: Georgi Djakov <djakov@kernel.org>
2024-03-05 22:56:52 +00:00
mutex_lock(&icc_bw_lock);
for (i = 0; i < path->num_nodes; i++) {
node = path->reqs[i].node;
hlist_del(&path->reqs[i].req_node);
if (!WARN_ON(!node->provider->users))
node->provider->users--;
}
interconnect: Don't access req_list while it's being manipulated The icc_lock mutex was split into separate icc_lock and icc_bw_lock mutexes in [1] to avoid lockdep splats. However, this didn't adequately protect access to icc_node::req_list. The icc_set_bw() function will eventually iterate over req_list while only holding icc_bw_lock, but req_list can be modified while only holding icc_lock. This causes races between icc_set_bw(), of_icc_get(), and icc_put(). Example A: CPU0 CPU1 ---- ---- icc_set_bw(path_a) mutex_lock(&icc_bw_lock); icc_put(path_b) mutex_lock(&icc_lock); aggregate_requests() hlist_for_each_entry(r, ... hlist_del(... <r = invalid pointer> Example B: CPU0 CPU1 ---- ---- icc_set_bw(path_a) mutex_lock(&icc_bw_lock); path_b = of_icc_get() of_icc_get_by_index() mutex_lock(&icc_lock); path_find() path_init() aggregate_requests() hlist_for_each_entry(r, ... hlist_add_head(... <r = invalid pointer> Fix this by ensuring icc_bw_lock is always held before manipulating icc_node::req_list. The additional places icc_bw_lock is held don't perform any memory allocations, so we should still be safe from the original lockdep splats that motivated the separate locks. [1] commit af42269c3523 ("interconnect: Fix locking for runpm vs reclaim") Signed-off-by: Mike Tipton <quic_mdtipton@quicinc.com> Fixes: af42269c3523 ("interconnect: Fix locking for runpm vs reclaim") Reviewed-by: Rob Clark <robdclark@chromium.org> Link: https://lore.kernel.org/r/20240305225652.22872-1-quic_mdtipton@quicinc.com Signed-off-by: Georgi Djakov <djakov@kernel.org>
2024-03-05 22:56:52 +00:00
mutex_unlock(&icc_bw_lock);
mutex_unlock(&icc_lock);
kfree_const(path->name);
kfree(path);
}
EXPORT_SYMBOL_GPL(icc_put);
static struct icc_node *icc_node_create_nolock(int id)
{
struct icc_node *node;
/* check if node already exists */
node = node_find(id);
if (node)
return node;
node = kzalloc(sizeof(*node), GFP_KERNEL);
if (!node)
return ERR_PTR(-ENOMEM);
id = idr_alloc(&icc_idr, node, id, id + 1, GFP_KERNEL);
if (id < 0) {
WARN(1, "%s: couldn't get idr\n", __func__);
kfree(node);
return ERR_PTR(id);
}
node->id = id;
return node;
}
/**
* icc_node_create() - create a node
* @id: node id
*
* Return: icc_node pointer on success, or ERR_PTR() on error
*/
struct icc_node *icc_node_create(int id)
{
struct icc_node *node;
mutex_lock(&icc_lock);
node = icc_node_create_nolock(id);
mutex_unlock(&icc_lock);
return node;
}
EXPORT_SYMBOL_GPL(icc_node_create);
/**
* icc_node_destroy() - destroy a node
* @id: node id
*/
void icc_node_destroy(int id)
{
struct icc_node *node;
mutex_lock(&icc_lock);
node = node_find(id);
if (node) {
idr_remove(&icc_idr, node->id);
WARN_ON(!hlist_empty(&node->req_list));
}
mutex_unlock(&icc_lock);
if (!node)
return;
kfree(node->links);
kfree(node);
}
EXPORT_SYMBOL_GPL(icc_node_destroy);
/**
* icc_link_create() - create a link between two nodes
* @node: source node id
* @dst_id: destination node id
*
* Create a link between two nodes. The nodes might belong to different
* interconnect providers and the @dst_id node might not exist (if the
* provider driver has not probed yet). So just create the @dst_id node
* and when the actual provider driver is probed, the rest of the node
* data is filled.
*
* Return: 0 on success, or an error code otherwise
*/
int icc_link_create(struct icc_node *node, const int dst_id)
{
struct icc_node *dst;
struct icc_node **new;
int ret = 0;
if (!node->provider)
return -EINVAL;
mutex_lock(&icc_lock);
dst = node_find(dst_id);
if (!dst) {
dst = icc_node_create_nolock(dst_id);
if (IS_ERR(dst)) {
ret = PTR_ERR(dst);
goto out;
}
}
new = krealloc(node->links,
(node->num_links + 1) * sizeof(*node->links),
GFP_KERNEL);
if (!new) {
ret = -ENOMEM;
goto out;
}
node->links = new;
node->links[node->num_links++] = dst;
out:
mutex_unlock(&icc_lock);
return ret;
}
EXPORT_SYMBOL_GPL(icc_link_create);
/**
* icc_node_add() - add interconnect node to interconnect provider
* @node: pointer to the interconnect node
* @provider: pointer to the interconnect provider
*/
void icc_node_add(struct icc_node *node, struct icc_provider *provider)
{
if (WARN_ON(node->provider))
return;
mutex_lock(&icc_lock);
interconnect: Fix locking for runpm vs reclaim For cases where icc_bw_set() can be called in callbaths that could deadlock against shrinker/reclaim, such as runpm resume, we need to decouple the icc locking. Introduce a new icc_bw_lock for cases where we need to serialize bw aggregation and update to decouple that from paths that require memory allocation such as node/link creation/ destruction. Fixes this lockdep splat: ====================================================== WARNING: possible circular locking dependency detected 6.2.0-rc8-debug+ #554 Not tainted ------------------------------------------------------ ring0/132 is trying to acquire lock: ffffff80871916d0 (&gmu->lock){+.+.}-{3:3}, at: a6xx_pm_resume+0xf0/0x234 but task is already holding lock: ffffffdb5aee57e8 (dma_fence_map){++++}-{0:0}, at: msm_job_run+0x68/0x150 which lock already depends on the new lock. the existing dependency chain (in reverse order) is: -> #4 (dma_fence_map){++++}-{0:0}: __dma_fence_might_wait+0x74/0xc0 dma_resv_lockdep+0x1f4/0x2f4 do_one_initcall+0x104/0x2bc kernel_init_freeable+0x344/0x34c kernel_init+0x30/0x134 ret_from_fork+0x10/0x20 -> #3 (mmu_notifier_invalidate_range_start){+.+.}-{0:0}: fs_reclaim_acquire+0x80/0xa8 slab_pre_alloc_hook.constprop.0+0x40/0x25c __kmem_cache_alloc_node+0x60/0x1cc __kmalloc+0xd8/0x100 topology_parse_cpu_capacity+0x8c/0x178 get_cpu_for_node+0x88/0xc4 parse_cluster+0x1b0/0x28c parse_cluster+0x8c/0x28c init_cpu_topology+0x168/0x188 smp_prepare_cpus+0x24/0xf8 kernel_init_freeable+0x18c/0x34c kernel_init+0x30/0x134 ret_from_fork+0x10/0x20 -> #2 (fs_reclaim){+.+.}-{0:0}: __fs_reclaim_acquire+0x3c/0x48 fs_reclaim_acquire+0x54/0xa8 slab_pre_alloc_hook.constprop.0+0x40/0x25c __kmem_cache_alloc_node+0x60/0x1cc __kmalloc+0xd8/0x100 kzalloc.constprop.0+0x14/0x20 icc_node_create_nolock+0x4c/0xc4 icc_node_create+0x38/0x58 qcom_icc_rpmh_probe+0x1b8/0x248 platform_probe+0x70/0xc4 really_probe+0x158/0x290 __driver_probe_device+0xc8/0xe0 driver_probe_device+0x44/0x100 __driver_attach+0xf8/0x108 bus_for_each_dev+0x78/0xc4 driver_attach+0x2c/0x38 bus_add_driver+0xd0/0x1d8 driver_register+0xbc/0xf8 __platform_driver_register+0x30/0x3c qnoc_driver_init+0x24/0x30 do_one_initcall+0x104/0x2bc kernel_init_freeable+0x344/0x34c kernel_init+0x30/0x134 ret_from_fork+0x10/0x20 -> #1 (icc_lock){+.+.}-{3:3}: __mutex_lock+0xcc/0x3c8 mutex_lock_nested+0x30/0x44 icc_set_bw+0x88/0x2b4 _set_opp_bw+0x8c/0xd8 _set_opp+0x19c/0x300 dev_pm_opp_set_opp+0x84/0x94 a6xx_gmu_resume+0x18c/0x804 a6xx_pm_resume+0xf8/0x234 adreno_runtime_resume+0x2c/0x38 pm_generic_runtime_resume+0x30/0x44 __rpm_callback+0x15c/0x174 rpm_callback+0x78/0x7c rpm_resume+0x318/0x524 __pm_runtime_resume+0x78/0xbc adreno_load_gpu+0xc4/0x17c msm_open+0x50/0x120 drm_file_alloc+0x17c/0x228 drm_open_helper+0x74/0x118 drm_open+0xa0/0x144 drm_stub_open+0xd4/0xe4 chrdev_open+0x1b8/0x1e4 do_dentry_open+0x2f8/0x38c vfs_open+0x34/0x40 path_openat+0x64c/0x7b4 do_filp_open+0x54/0xc4 do_sys_openat2+0x9c/0x100 do_sys_open+0x50/0x7c __arm64_sys_openat+0x28/0x34 invoke_syscall+0x8c/0x128 el0_svc_common.constprop.0+0xa0/0x11c do_el0_svc+0xac/0xbc el0_svc+0x48/0xa0 el0t_64_sync_handler+0xac/0x13c el0t_64_sync+0x190/0x194 -> #0 (&gmu->lock){+.+.}-{3:3}: __lock_acquire+0xe00/0x1060 lock_acquire+0x1e0/0x2f8 __mutex_lock+0xcc/0x3c8 mutex_lock_nested+0x30/0x44 a6xx_pm_resume+0xf0/0x234 adreno_runtime_resume+0x2c/0x38 pm_generic_runtime_resume+0x30/0x44 __rpm_callback+0x15c/0x174 rpm_callback+0x78/0x7c rpm_resume+0x318/0x524 __pm_runtime_resume+0x78/0xbc pm_runtime_get_sync.isra.0+0x14/0x20 msm_gpu_submit+0x58/0x178 msm_job_run+0x78/0x150 drm_sched_main+0x290/0x370 kthread+0xf0/0x100 ret_from_fork+0x10/0x20 other info that might help us debug this: Chain exists of: &gmu->lock --> mmu_notifier_invalidate_range_start --> dma_fence_map Possible unsafe locking scenario: CPU0 CPU1 ---- ---- lock(dma_fence_map); lock(mmu_notifier_invalidate_range_start); lock(dma_fence_map); lock(&gmu->lock); *** DEADLOCK *** 2 locks held by ring0/132: #0: ffffff8087191170 (&gpu->lock){+.+.}-{3:3}, at: msm_job_run+0x64/0x150 #1: ffffffdb5aee57e8 (dma_fence_map){++++}-{0:0}, at: msm_job_run+0x68/0x150 stack backtrace: CPU: 7 PID: 132 Comm: ring0 Not tainted 6.2.0-rc8-debug+ #554 Hardware name: Google Lazor (rev1 - 2) with LTE (DT) Call trace: dump_backtrace.part.0+0xb4/0xf8 show_stack+0x20/0x38 dump_stack_lvl+0x9c/0xd0 dump_stack+0x18/0x34 print_circular_bug+0x1b4/0x1f0 check_noncircular+0x78/0xac __lock_acquire+0xe00/0x1060 lock_acquire+0x1e0/0x2f8 __mutex_lock+0xcc/0x3c8 mutex_lock_nested+0x30/0x44 a6xx_pm_resume+0xf0/0x234 adreno_runtime_resume+0x2c/0x38 pm_generic_runtime_resume+0x30/0x44 __rpm_callback+0x15c/0x174 rpm_callback+0x78/0x7c rpm_resume+0x318/0x524 __pm_runtime_resume+0x78/0xbc pm_runtime_get_sync.isra.0+0x14/0x20 msm_gpu_submit+0x58/0x178 msm_job_run+0x78/0x150 drm_sched_main+0x290/0x370 kthread+0xf0/0x100 ret_from_fork+0x10/0x20 Signed-off-by: Rob Clark <robdclark@chromium.org> Link: https://lore.kernel.org/r/20230807171148.210181-7-robdclark@gmail.com Signed-off-by: Georgi Djakov <djakov@kernel.org>
2023-08-07 17:11:40 +00:00
mutex_lock(&icc_bw_lock);
node->provider = provider;
list_add_tail(&node->node_list, &provider->nodes);
/* get the initial bandwidth values and sync them with hardware */
if (provider->get_bw) {
provider->get_bw(node, &node->init_avg, &node->init_peak);
} else {
node->init_avg = INT_MAX;
node->init_peak = INT_MAX;
}
node->avg_bw = node->init_avg;
node->peak_bw = node->init_peak;
if (node->avg_bw || node->peak_bw) {
if (provider->pre_aggregate)
provider->pre_aggregate(node);
if (provider->aggregate)
provider->aggregate(node, 0, node->init_avg, node->init_peak,
&node->avg_bw, &node->peak_bw);
if (provider->set)
provider->set(node, node);
}
node->avg_bw = 0;
node->peak_bw = 0;
interconnect: Fix locking for runpm vs reclaim For cases where icc_bw_set() can be called in callbaths that could deadlock against shrinker/reclaim, such as runpm resume, we need to decouple the icc locking. Introduce a new icc_bw_lock for cases where we need to serialize bw aggregation and update to decouple that from paths that require memory allocation such as node/link creation/ destruction. Fixes this lockdep splat: ====================================================== WARNING: possible circular locking dependency detected 6.2.0-rc8-debug+ #554 Not tainted ------------------------------------------------------ ring0/132 is trying to acquire lock: ffffff80871916d0 (&gmu->lock){+.+.}-{3:3}, at: a6xx_pm_resume+0xf0/0x234 but task is already holding lock: ffffffdb5aee57e8 (dma_fence_map){++++}-{0:0}, at: msm_job_run+0x68/0x150 which lock already depends on the new lock. the existing dependency chain (in reverse order) is: -> #4 (dma_fence_map){++++}-{0:0}: __dma_fence_might_wait+0x74/0xc0 dma_resv_lockdep+0x1f4/0x2f4 do_one_initcall+0x104/0x2bc kernel_init_freeable+0x344/0x34c kernel_init+0x30/0x134 ret_from_fork+0x10/0x20 -> #3 (mmu_notifier_invalidate_range_start){+.+.}-{0:0}: fs_reclaim_acquire+0x80/0xa8 slab_pre_alloc_hook.constprop.0+0x40/0x25c __kmem_cache_alloc_node+0x60/0x1cc __kmalloc+0xd8/0x100 topology_parse_cpu_capacity+0x8c/0x178 get_cpu_for_node+0x88/0xc4 parse_cluster+0x1b0/0x28c parse_cluster+0x8c/0x28c init_cpu_topology+0x168/0x188 smp_prepare_cpus+0x24/0xf8 kernel_init_freeable+0x18c/0x34c kernel_init+0x30/0x134 ret_from_fork+0x10/0x20 -> #2 (fs_reclaim){+.+.}-{0:0}: __fs_reclaim_acquire+0x3c/0x48 fs_reclaim_acquire+0x54/0xa8 slab_pre_alloc_hook.constprop.0+0x40/0x25c __kmem_cache_alloc_node+0x60/0x1cc __kmalloc+0xd8/0x100 kzalloc.constprop.0+0x14/0x20 icc_node_create_nolock+0x4c/0xc4 icc_node_create+0x38/0x58 qcom_icc_rpmh_probe+0x1b8/0x248 platform_probe+0x70/0xc4 really_probe+0x158/0x290 __driver_probe_device+0xc8/0xe0 driver_probe_device+0x44/0x100 __driver_attach+0xf8/0x108 bus_for_each_dev+0x78/0xc4 driver_attach+0x2c/0x38 bus_add_driver+0xd0/0x1d8 driver_register+0xbc/0xf8 __platform_driver_register+0x30/0x3c qnoc_driver_init+0x24/0x30 do_one_initcall+0x104/0x2bc kernel_init_freeable+0x344/0x34c kernel_init+0x30/0x134 ret_from_fork+0x10/0x20 -> #1 (icc_lock){+.+.}-{3:3}: __mutex_lock+0xcc/0x3c8 mutex_lock_nested+0x30/0x44 icc_set_bw+0x88/0x2b4 _set_opp_bw+0x8c/0xd8 _set_opp+0x19c/0x300 dev_pm_opp_set_opp+0x84/0x94 a6xx_gmu_resume+0x18c/0x804 a6xx_pm_resume+0xf8/0x234 adreno_runtime_resume+0x2c/0x38 pm_generic_runtime_resume+0x30/0x44 __rpm_callback+0x15c/0x174 rpm_callback+0x78/0x7c rpm_resume+0x318/0x524 __pm_runtime_resume+0x78/0xbc adreno_load_gpu+0xc4/0x17c msm_open+0x50/0x120 drm_file_alloc+0x17c/0x228 drm_open_helper+0x74/0x118 drm_open+0xa0/0x144 drm_stub_open+0xd4/0xe4 chrdev_open+0x1b8/0x1e4 do_dentry_open+0x2f8/0x38c vfs_open+0x34/0x40 path_openat+0x64c/0x7b4 do_filp_open+0x54/0xc4 do_sys_openat2+0x9c/0x100 do_sys_open+0x50/0x7c __arm64_sys_openat+0x28/0x34 invoke_syscall+0x8c/0x128 el0_svc_common.constprop.0+0xa0/0x11c do_el0_svc+0xac/0xbc el0_svc+0x48/0xa0 el0t_64_sync_handler+0xac/0x13c el0t_64_sync+0x190/0x194 -> #0 (&gmu->lock){+.+.}-{3:3}: __lock_acquire+0xe00/0x1060 lock_acquire+0x1e0/0x2f8 __mutex_lock+0xcc/0x3c8 mutex_lock_nested+0x30/0x44 a6xx_pm_resume+0xf0/0x234 adreno_runtime_resume+0x2c/0x38 pm_generic_runtime_resume+0x30/0x44 __rpm_callback+0x15c/0x174 rpm_callback+0x78/0x7c rpm_resume+0x318/0x524 __pm_runtime_resume+0x78/0xbc pm_runtime_get_sync.isra.0+0x14/0x20 msm_gpu_submit+0x58/0x178 msm_job_run+0x78/0x150 drm_sched_main+0x290/0x370 kthread+0xf0/0x100 ret_from_fork+0x10/0x20 other info that might help us debug this: Chain exists of: &gmu->lock --> mmu_notifier_invalidate_range_start --> dma_fence_map Possible unsafe locking scenario: CPU0 CPU1 ---- ---- lock(dma_fence_map); lock(mmu_notifier_invalidate_range_start); lock(dma_fence_map); lock(&gmu->lock); *** DEADLOCK *** 2 locks held by ring0/132: #0: ffffff8087191170 (&gpu->lock){+.+.}-{3:3}, at: msm_job_run+0x64/0x150 #1: ffffffdb5aee57e8 (dma_fence_map){++++}-{0:0}, at: msm_job_run+0x68/0x150 stack backtrace: CPU: 7 PID: 132 Comm: ring0 Not tainted 6.2.0-rc8-debug+ #554 Hardware name: Google Lazor (rev1 - 2) with LTE (DT) Call trace: dump_backtrace.part.0+0xb4/0xf8 show_stack+0x20/0x38 dump_stack_lvl+0x9c/0xd0 dump_stack+0x18/0x34 print_circular_bug+0x1b4/0x1f0 check_noncircular+0x78/0xac __lock_acquire+0xe00/0x1060 lock_acquire+0x1e0/0x2f8 __mutex_lock+0xcc/0x3c8 mutex_lock_nested+0x30/0x44 a6xx_pm_resume+0xf0/0x234 adreno_runtime_resume+0x2c/0x38 pm_generic_runtime_resume+0x30/0x44 __rpm_callback+0x15c/0x174 rpm_callback+0x78/0x7c rpm_resume+0x318/0x524 __pm_runtime_resume+0x78/0xbc pm_runtime_get_sync.isra.0+0x14/0x20 msm_gpu_submit+0x58/0x178 msm_job_run+0x78/0x150 drm_sched_main+0x290/0x370 kthread+0xf0/0x100 ret_from_fork+0x10/0x20 Signed-off-by: Rob Clark <robdclark@chromium.org> Link: https://lore.kernel.org/r/20230807171148.210181-7-robdclark@gmail.com Signed-off-by: Georgi Djakov <djakov@kernel.org>
2023-08-07 17:11:40 +00:00
mutex_unlock(&icc_bw_lock);
mutex_unlock(&icc_lock);
}
EXPORT_SYMBOL_GPL(icc_node_add);
/**
* icc_node_del() - delete interconnect node from interconnect provider
* @node: pointer to the interconnect node
*/
void icc_node_del(struct icc_node *node)
{
mutex_lock(&icc_lock);
list_del(&node->node_list);
mutex_unlock(&icc_lock);
}
EXPORT_SYMBOL_GPL(icc_node_del);
/**
* icc_nodes_remove() - remove all previously added nodes from provider
* @provider: the interconnect provider we are removing nodes from
*
* Return: 0 on success, or an error code otherwise
*/
int icc_nodes_remove(struct icc_provider *provider)
{
struct icc_node *n, *tmp;
if (WARN_ON(IS_ERR_OR_NULL(provider)))
return -EINVAL;
list_for_each_entry_safe_reverse(n, tmp, &provider->nodes, node_list) {
icc_node_del(n);
icc_node_destroy(n->id);
}
return 0;
}
EXPORT_SYMBOL_GPL(icc_nodes_remove);
/**
* icc_provider_init() - initialize a new interconnect provider
* @provider: the interconnect provider to initialize
*
* Must be called before adding nodes to the provider.
*/
void icc_provider_init(struct icc_provider *provider)
{
WARN_ON(!provider->set);
INIT_LIST_HEAD(&provider->nodes);
}
EXPORT_SYMBOL_GPL(icc_provider_init);
/**
* icc_provider_register() - register a new interconnect provider
* @provider: the interconnect provider to register
*
* Return: 0 on success, or an error code otherwise
*/
int icc_provider_register(struct icc_provider *provider)
{
if (WARN_ON(!provider->xlate && !provider->xlate_extended))
return -EINVAL;
mutex_lock(&icc_lock);
list_add_tail(&provider->provider_list, &icc_providers);
mutex_unlock(&icc_lock);
dev_dbg(provider->dev, "interconnect provider registered\n");
return 0;
}
EXPORT_SYMBOL_GPL(icc_provider_register);
/**
* icc_provider_deregister() - deregister an interconnect provider
* @provider: the interconnect provider to deregister
*/
void icc_provider_deregister(struct icc_provider *provider)
{
mutex_lock(&icc_lock);
WARN_ON(provider->users);
list_del(&provider->provider_list);
mutex_unlock(&icc_lock);
}
EXPORT_SYMBOL_GPL(icc_provider_deregister);
static const struct of_device_id __maybe_unused ignore_list[] = {
{ .compatible = "qcom,sc7180-ipa-virt" },
{ .compatible = "qcom,sc8180x-ipa-virt" },
{ .compatible = "qcom,sdx55-ipa-virt" },
{ .compatible = "qcom,sm8150-ipa-virt" },
{ .compatible = "qcom,sm8250-ipa-virt" },
{}
};
static int of_count_icc_providers(struct device_node *np)
{
struct device_node *child;
int count = 0;
for_each_available_child_of_node(np, child) {
if (of_property_read_bool(child, "#interconnect-cells") &&
likely(!of_match_node(ignore_list, child)))
count++;
count += of_count_icc_providers(child);
}
return count;
}
void icc_sync_state(struct device *dev)
{
struct icc_provider *p;
struct icc_node *n;
static int count;
count++;
if (count < providers_count)
return;
mutex_lock(&icc_lock);
interconnect: Fix locking for runpm vs reclaim For cases where icc_bw_set() can be called in callbaths that could deadlock against shrinker/reclaim, such as runpm resume, we need to decouple the icc locking. Introduce a new icc_bw_lock for cases where we need to serialize bw aggregation and update to decouple that from paths that require memory allocation such as node/link creation/ destruction. Fixes this lockdep splat: ====================================================== WARNING: possible circular locking dependency detected 6.2.0-rc8-debug+ #554 Not tainted ------------------------------------------------------ ring0/132 is trying to acquire lock: ffffff80871916d0 (&gmu->lock){+.+.}-{3:3}, at: a6xx_pm_resume+0xf0/0x234 but task is already holding lock: ffffffdb5aee57e8 (dma_fence_map){++++}-{0:0}, at: msm_job_run+0x68/0x150 which lock already depends on the new lock. the existing dependency chain (in reverse order) is: -> #4 (dma_fence_map){++++}-{0:0}: __dma_fence_might_wait+0x74/0xc0 dma_resv_lockdep+0x1f4/0x2f4 do_one_initcall+0x104/0x2bc kernel_init_freeable+0x344/0x34c kernel_init+0x30/0x134 ret_from_fork+0x10/0x20 -> #3 (mmu_notifier_invalidate_range_start){+.+.}-{0:0}: fs_reclaim_acquire+0x80/0xa8 slab_pre_alloc_hook.constprop.0+0x40/0x25c __kmem_cache_alloc_node+0x60/0x1cc __kmalloc+0xd8/0x100 topology_parse_cpu_capacity+0x8c/0x178 get_cpu_for_node+0x88/0xc4 parse_cluster+0x1b0/0x28c parse_cluster+0x8c/0x28c init_cpu_topology+0x168/0x188 smp_prepare_cpus+0x24/0xf8 kernel_init_freeable+0x18c/0x34c kernel_init+0x30/0x134 ret_from_fork+0x10/0x20 -> #2 (fs_reclaim){+.+.}-{0:0}: __fs_reclaim_acquire+0x3c/0x48 fs_reclaim_acquire+0x54/0xa8 slab_pre_alloc_hook.constprop.0+0x40/0x25c __kmem_cache_alloc_node+0x60/0x1cc __kmalloc+0xd8/0x100 kzalloc.constprop.0+0x14/0x20 icc_node_create_nolock+0x4c/0xc4 icc_node_create+0x38/0x58 qcom_icc_rpmh_probe+0x1b8/0x248 platform_probe+0x70/0xc4 really_probe+0x158/0x290 __driver_probe_device+0xc8/0xe0 driver_probe_device+0x44/0x100 __driver_attach+0xf8/0x108 bus_for_each_dev+0x78/0xc4 driver_attach+0x2c/0x38 bus_add_driver+0xd0/0x1d8 driver_register+0xbc/0xf8 __platform_driver_register+0x30/0x3c qnoc_driver_init+0x24/0x30 do_one_initcall+0x104/0x2bc kernel_init_freeable+0x344/0x34c kernel_init+0x30/0x134 ret_from_fork+0x10/0x20 -> #1 (icc_lock){+.+.}-{3:3}: __mutex_lock+0xcc/0x3c8 mutex_lock_nested+0x30/0x44 icc_set_bw+0x88/0x2b4 _set_opp_bw+0x8c/0xd8 _set_opp+0x19c/0x300 dev_pm_opp_set_opp+0x84/0x94 a6xx_gmu_resume+0x18c/0x804 a6xx_pm_resume+0xf8/0x234 adreno_runtime_resume+0x2c/0x38 pm_generic_runtime_resume+0x30/0x44 __rpm_callback+0x15c/0x174 rpm_callback+0x78/0x7c rpm_resume+0x318/0x524 __pm_runtime_resume+0x78/0xbc adreno_load_gpu+0xc4/0x17c msm_open+0x50/0x120 drm_file_alloc+0x17c/0x228 drm_open_helper+0x74/0x118 drm_open+0xa0/0x144 drm_stub_open+0xd4/0xe4 chrdev_open+0x1b8/0x1e4 do_dentry_open+0x2f8/0x38c vfs_open+0x34/0x40 path_openat+0x64c/0x7b4 do_filp_open+0x54/0xc4 do_sys_openat2+0x9c/0x100 do_sys_open+0x50/0x7c __arm64_sys_openat+0x28/0x34 invoke_syscall+0x8c/0x128 el0_svc_common.constprop.0+0xa0/0x11c do_el0_svc+0xac/0xbc el0_svc+0x48/0xa0 el0t_64_sync_handler+0xac/0x13c el0t_64_sync+0x190/0x194 -> #0 (&gmu->lock){+.+.}-{3:3}: __lock_acquire+0xe00/0x1060 lock_acquire+0x1e0/0x2f8 __mutex_lock+0xcc/0x3c8 mutex_lock_nested+0x30/0x44 a6xx_pm_resume+0xf0/0x234 adreno_runtime_resume+0x2c/0x38 pm_generic_runtime_resume+0x30/0x44 __rpm_callback+0x15c/0x174 rpm_callback+0x78/0x7c rpm_resume+0x318/0x524 __pm_runtime_resume+0x78/0xbc pm_runtime_get_sync.isra.0+0x14/0x20 msm_gpu_submit+0x58/0x178 msm_job_run+0x78/0x150 drm_sched_main+0x290/0x370 kthread+0xf0/0x100 ret_from_fork+0x10/0x20 other info that might help us debug this: Chain exists of: &gmu->lock --> mmu_notifier_invalidate_range_start --> dma_fence_map Possible unsafe locking scenario: CPU0 CPU1 ---- ---- lock(dma_fence_map); lock(mmu_notifier_invalidate_range_start); lock(dma_fence_map); lock(&gmu->lock); *** DEADLOCK *** 2 locks held by ring0/132: #0: ffffff8087191170 (&gpu->lock){+.+.}-{3:3}, at: msm_job_run+0x64/0x150 #1: ffffffdb5aee57e8 (dma_fence_map){++++}-{0:0}, at: msm_job_run+0x68/0x150 stack backtrace: CPU: 7 PID: 132 Comm: ring0 Not tainted 6.2.0-rc8-debug+ #554 Hardware name: Google Lazor (rev1 - 2) with LTE (DT) Call trace: dump_backtrace.part.0+0xb4/0xf8 show_stack+0x20/0x38 dump_stack_lvl+0x9c/0xd0 dump_stack+0x18/0x34 print_circular_bug+0x1b4/0x1f0 check_noncircular+0x78/0xac __lock_acquire+0xe00/0x1060 lock_acquire+0x1e0/0x2f8 __mutex_lock+0xcc/0x3c8 mutex_lock_nested+0x30/0x44 a6xx_pm_resume+0xf0/0x234 adreno_runtime_resume+0x2c/0x38 pm_generic_runtime_resume+0x30/0x44 __rpm_callback+0x15c/0x174 rpm_callback+0x78/0x7c rpm_resume+0x318/0x524 __pm_runtime_resume+0x78/0xbc pm_runtime_get_sync.isra.0+0x14/0x20 msm_gpu_submit+0x58/0x178 msm_job_run+0x78/0x150 drm_sched_main+0x290/0x370 kthread+0xf0/0x100 ret_from_fork+0x10/0x20 Signed-off-by: Rob Clark <robdclark@chromium.org> Link: https://lore.kernel.org/r/20230807171148.210181-7-robdclark@gmail.com Signed-off-by: Georgi Djakov <djakov@kernel.org>
2023-08-07 17:11:40 +00:00
mutex_lock(&icc_bw_lock);
synced_state = true;
list_for_each_entry(p, &icc_providers, provider_list) {
dev_dbg(p->dev, "interconnect provider is in synced state\n");
list_for_each_entry(n, &p->nodes, node_list) {
if (n->init_avg || n->init_peak) {
n->init_avg = 0;
n->init_peak = 0;
aggregate_requests(n);
p->set(n, n);
}
}
}
mutex_unlock(&icc_bw_lock);
mutex_unlock(&icc_lock);
}
EXPORT_SYMBOL_GPL(icc_sync_state);
static int __init icc_init(void)
{
struct device_node *root;
/* Teach lockdep about lock ordering wrt. shrinker: */
fs_reclaim_acquire(GFP_KERNEL);
might_lock(&icc_bw_lock);
fs_reclaim_release(GFP_KERNEL);
root = of_find_node_by_path("/");
providers_count = of_count_icc_providers(root);
of_node_put(root);
icc_debugfs_dir = debugfs_create_dir("interconnect", NULL);
debugfs_create_file("interconnect_summary", 0444,
icc_debugfs_dir, NULL, &icc_summary_fops);
debugfs_create_file("interconnect_graph", 0444,
icc_debugfs_dir, NULL, &icc_graph_fops);
icc_debugfs_client_init(icc_debugfs_dir);
return 0;
}
device_initcall(icc_init);